Skip to main content

Table 2 AUCs of prediction models based on different training sample sizes and multiple sampling times

From: Factors affecting the performance of brain arteriovenous malformation rupture prediction models

Algorithm

LR

RF

p value

Training sample size/test sample size

140/213

175/178

210/143

245/108

280/73

140/213

175/178

210/143

245/108

280/73

Sampling times_1-1 [mean (SD)]

0.67 (NA)

0.67 (NA)

0.70 (NA)

0.61 (NA)

0.75 (NA)

0.63 (NA)

0.76 (NA)

0.65 (NA)

0.62 (NA)

0.76 (NA)

<ā€‰0.001

Sampling times_1-2 [mean (SD)]

0.69 (NA)

0.73 (NA)

0.75 (NA)

0.68 (NA)

0.80 (NA)

0.69 (NA)

0.70 (NA)

0.77 (NA)

0.69 (NA)

0.73 (NA)

Sampling times_1-3 [mean (SD)]

0.70 (NA)

0.69 (NA)

0.69 (NA)

0.77 (NA)

0.57 (NA)

0.67 (NA)

0.70 (NA)

0.66 (NA)

0.64 (NA)

0.58 (NA)

Sampling times_1-4 [mean (SD)]

0.63 (NA)

0.68 (NA)

0.71 (NA)

0.68 (NA)

0.68 (NA)

0.70 (NA)

0.68 (NA)

0.66 (NA)

0.72 (NA)

0.72 (NA)

Sampling times_5 [mean (SD)]

0.69 (0.028)

0.71 (0.025)

0.71 (0.041)

0.70 (0.057)

0.71 (0.058)

0.66 (0.024)

0.68 (0.028)

0.71 (0.037)

0.66 (0.050)

0.63 (0.031)

Sampling times_10 [mean (SD)]

0.70 (0.021)

0.70 (0.019)

0.70 (0.091)

0.71 (0.038)

0.71 (0.049)

0.68 (0.035)

0.67 (0.030)

0.68 (0.042)

0.68 (0.044)

0.67 (0.020)

Sampling times_50 [mean (SD)]

0.71 (0.029)

0.70 (0.036)

0.70 (0.040)

0.72 (0.042)

0.72 (0.061)

0.68 (0.033)

0.69 (0.038)

0.68 (0.035)

0.68 (0.043)

0.69 (0.058)

Sampling times_100 [mean (SD)]

0.70 (0.029)

0.70 (0.032)

0.71 (0.040)

0.71 (0.045)

0.71 (0.063)

0.68 (0.028)

0.69 (0.036)

0.68 (0.044)

0.68 (0.045)

0.68 (0.061)

Sampling times_300 [mean (SD)]

0.70 (0.033)

0.71 (0.032)

0.70 (0.038)

0.71 (0.046)

0.71 (0.062)

0.68 (0.027)

0.69 (0.034)

0.68 (0.038)

0.68 (0.043)

0.67 (0.059)

Sampling times_600 [mean (SD)]

0.70 (0.033)

0.71 (0.033)

0.71 (0.038)

0.71 (0.047)

0.71 (0.061)

0.68 (0.027)

0.69 (0.032)

0.68 (0.037)

0.68 (0.047)

0.68 (0.062)

Sampling times_1200 [mean (SD)]

0.70 (0.033)

0.70 (0.033)

0.71 (0.038)

0.71 (0.044)

0.71 (0.061)

0.68 (0.028)

0.68 (0.033)

0.68 (0.039)

0.68 (0.044)

0.68 (0.059)

Sampling times_2100 [mean (SD)]

0.70 (0.033)

0.70 (0.033)

0.71 (0.037)

0.71 (0.046)

0.71 (0.060)

0.68 (0.027)

0.69 (0.032)

0.68 (0.039)

0.68 (0.047)

0.68 (0.059)

  1. This table corresponds to Fig.Ā 1
  2. AUC area under the curve, LR logistic regression, RF random forest, SD standard deviations